4 Experts on the Future of London Tech
London has established itself as a global tech powerhouse. How will the city make the most of this movement to ensure prosperity for businesses and talent?
Last month, BrainStation hosted Demo Day for the Summer 2019 graduates of the Web Development, UX Design, and Data Science Diploma programs. Take a look at some of the work these students were able to accomplish in just 12 weeks.
High
James Kachan, UX Design Graduate
High is a specialty journal that helps cannabis users record, monitor, and improve their experience with different cannabis strains. The app aids in tracking cannabis usage by strain, method of consumption, and resulting effects. It also provides a space to explore thoughts and experiences through images and writing. Visit James’ portfolio to keep up with his work.
Subtle
Kenneth Namgung, Web Development Graduate
Subtle is a machine-learning application that analyzes poses in photographs to provide a summary of repetitive gestures. During a photoshoot, the photographer is able to categorize results based on the type of shot they’ve captured.
Manga Face Detector
Mengyao Jia, Data Science Graduate
Mengyao’s project employes R-CNN to retrieve anime characters from the Manga109 database. With this raw data, she was able to use machine learning to accurately identify the faces of manga characters. See the project in action below or take a look at her work on Github.
Shape
Shape is a mobile platform powered by Artificial Intelligence to help find your best fit when shopping online. By referencing the measurements of brands you’re already familiar with — Shape suggests new and unfamiliar products to enhance the consumer’s shopping experience. Take a look at Joel’s complete portfolio.
Hire or Not?
Sam Joshva, Data Science Graduate
Sam strives to unravel the mystery of hiring in big companies. Have you ever wondered why you haven’t gotten a call-back after sending a resume? This program works to predict a candidate’s success based on the contents of their resume.
Healthwise
Mosope Adebowale, UX Design Graduate
Healthwise is a virtual healthcare platform that provides an easier way to connect patients with doctors in Nigeria, one of the most developed African countries in one of the worst states concerning healthcare. The service provides online consultation through voice, video or chat. See Mosope’s full profile.
Facial Expression Recognizer
Arash Tavassoli, Data Science Graduate
Arash created a deep-learning system to recognize emotional expressions from images of human faces. The Facial Expression Recognition represents an Image Classification problem within the wider field of Computer Vision. In this project, a machine learning system is developed to recognize emotional expressions (ex. happiness, sadness, or surprise) from images of human faces using a Convolutional Neural Network (CNN) in TensorFlow and OpenCV. Arash has completed an array of other work that you can see on his full portfolio.
Streetza
Omar Khan, Web Development Graduate
“The year is 2135. It’s a dystopian, awful future and some time in the past century, pizza was banned.” Streetza develops a map of pizza parlours along your cycling route to help you find the underground pizza market near you. As an avid biker, Omar Khan wanted to work with the Strava API, which he uses to map his rides. He used React for the front-end client and a Node/Express server to handle backend requests.
Plural.AI
Plural.AI is an interactive news aggregation app that makes it easy for millennials to stay informed on all sides of politically contentious issues while building empathy for different viewpoints. Powered by artificial intelligence, Plural.AI’s robo-journalist pulls the latest news stories from all over the web and sorts them into the app based on its bias detection algorithm, using the data to write neutral stories that are updated constantly in real time.
BigBrainBeat
Vanin Ferrall, Data Science Graduate
Vanin trained a convolution neural network to predict the BPM of four second slices of music. From this prediction it would then infer where the beats are within this slice. The result: The BigBrainBeat program takes 4 second slices of music and guesses the musical tempo (beats per minute) in the slice. See the work behind the project on Github or view Vanin’s full portfolio.
Peak
Sumit Sekhri, UX Design Graduate
Peak makes it easy to find the perfect ski mountain. Sumit’s app will show you the nearest mountains, their rating, recent snowfall, and the runs available.
Cancer Detection
Romina Caraba, Data Science Graduate
From genetic therapy to detection of various diseases, AI has revolutionized medical research. Recently, health researchers have applied AI to one of the most pressing health concerns of the century: cancer. Breast cancer is the second most common cancer in women. Romina created a cancer detection project that trains an AI model to detect cancer tissue in microscopic images. Early detection of metastatic tissue could potentially support health providers in developing swift diagnosis for patients, possibly saving their lives. See Romina’s Github for more details on her work.
Site Assistant
Dale Shlass, Web Development Graduate
Site Assistant is an application for both Construction Supervisors and Site Engineers. The app allows the user to take field notes and photos for their reports on the construction site. Once reports are created, they are automatically distributed by email to the project clients. Take a look at Dale’s portfolio, or see more of his work on Github.
Wardrobe Aid
Laura Bresson, Data Science Graduate
During her time at BrainStation, Laura discovered her love of Machine Learning and Deep Learning. She used a neural network to classify fashion product images that, after training, achieved an accuracy of over 99% in classifying bags, shoes, top wear and bottom wear. Connect with Laura on LinkedIn and see more of her projects on GitHub.
Bike Share Use
Philip Piltch, Data Science Graduate
Philip conducted an analysis of Toronto Bike Share, looking at how trip volume changes over time. An Auto-Regression model was used to predict trip volume for three different stations, one with the most number of trips, one with the least number of trips and one with the most round trips. Data was obtained from the City of Toronto Open Data portal. See Phil’s process on his Github.
Minding our Madness
Nattie blurs the lines of traditional statistical analysis and machine learning in her capstone project by using a combination of techniques from both disciplines. She explore employees’ comfort levels in discussing mental health issues at workplaces and proposes strategies to improve support for mental wellness in the tech industry. The visualization below represents answers to the question: “Briefly describe what you think the industry as a whole and/or employees could do to improve mental health support for employees.” Connect with Nattie here to learn more about her project.
